[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAHp75Vchgj=GXgY4uoDQ3YD7wqTbwWmm+BM=UW1MHt3qxUwEvw@mail.gmail.com>
Date: Wed, 3 Dec 2025 11:18:48 +0200
From: Andy Shevchenko <andy.shevchenko@...il.com>
To: Crescent Hsieh <crescentcy.hsieh@...a.com>
Cc: gregkh@...uxfoundation.org, jirislaby@...nel.org,
ilpo.jarvinen@...ux.intel.com, linux-kernel@...r.kernel.org,
linux-serial@...r.kernel.org
Subject: Re: [PATCH v1 09/31] serial: 8250: split 8250_mxpcie from 8250_pci
On Wed, Dec 3, 2025 at 4:54 AM Crescent Hsieh <crescentcy.hsieh@...a.com> wrote:
>
> On Sun, Nov 30, 2025 at 09:22:20PM +0200, Andy Shevchenko wrote:
> > On Sun, Nov 30, 2025 at 12:44 PM Crescent Hsieh
> > <crescentcy.hsieh@...a.com> wrote:
> > > +#include <linux/module.h>
> > > +#include <linux/pci.h>
> >
> > This needs more headers to be included
> >
> > bitfield.h
> > bits.h
> > dev_printk.h
> > device/devres.h
> > ioport.h
> > types.h
>
> May I ask a clarification: my understanding was to include headers only
> when the corresponding symbols are directly used.
Yes, and that's what I have checked (I spent some time reading the
whole patch, but I might even miss something, so this list might be
not comprehensive).
> Is the intention here to avoid relying on implicit includes and to make
> the driver’s dependencies explicit?
Yes.
--
With Best Regards,
Andy Shevchenko
Powered by blists - more mailing lists